Age Variation in Bitter Taste Perception in Relation to the Tas2r38 Taste Receptor Phenotype

Rossella Negri, Andrea Smarrazzo, Martina Galatola, Antonietta Maio, Paola Iaccarino Idelson Maura Sticco , Carmen Biongiovanni , Adriana Franzese , Luigi Greco, Davide Risso, Gabriella Morini 1 Department of Translational Medicine, Pediatrics, European Laboratory for Food Induced Diseases, University of Naples Federico II, Italy. 2 Lab. of Molecular Anthropology & Centre for Genome Biology, Dept. BIGEA, University of Bologna, Italy 3 University of Gastronomic Sciences, Pollenzo, Bra (CN), Italy.


Introduction
Recent discoveries related to the molecular domain of human taste have opened new avenues for the study of the relationship between inherited factors and food behaviour [1,2]. This is particularly important In fact, they much prefer sweet (energy) and reject bitter (potentially toxic) foods; consequently, they tend to avoid low energy-density foods like vegetables and fruit, and favour high-energy density-rich food such as sugars and fats [3] To date, the most studied genotype-phenotype association related to taste is the one of Phenylthiocarbamide (PTC) and the related compound 6n-propylthiouracil (PROP), both not found in nature, which sensitivity closely related to common genetic variants in the TAS2R38 gene located on chromosome [7].
PROP bitter taste sensitivity appears to be a marker of a wide variety of factors that condition food choices. Indeed, hypersensitivity to propylthiouracil (PROP) is associated to heightened responses to sweeteners [5]and salt [6], as well as to such chemesthetic [7,8] and somatosensory [9,10] sensations as the feel of astringency and the hotness of fats in the mouth [11]. Therefore, the PROP bitter taste phenotype is considered a general marker of oral sensitivity that is able to influence food selection and body weight [12][13][14][15][16][17]; however, these results were not confirmed by other studies [18,19].
Individuals are usually classified as being PROP bitter insensitive, sensitive or super sensitive according to the perceived intensity of the PROP solutions [4]. In this context, the salivary protein gustin (CAVI) that has been implicated in taste bud growth and maintenance was implicated in the control of PROP phenotype [21,22].
We previously confirmed the relationship between the TAS2R38 genotype and the PROP taste phenotype in children and adults, and observed a difference in this relationship between children and adults carrying the same genotype [20]. The aim of the present study was to examine systematically in a crosssection study how TAS2R38 and CAVI polymorphisms affect bitter taste perception and food preference in relation to age and sex. We also explored the feeding behaviour of children in relation to their mother's food preferences in order to stratify for familial environment and shared or unshared genomic profile.  In case of dubious expressions, the child was retested and his/her mother helped to interpret the child's expression. Based on the score, subjects were classified as non-tasters (score 0-2), medium tasters (score 3-4), and super tasters (score 5-6). The score is based on the two PROP solutions.

Study
For quality control, a random sample of 30 adults was tested for threshold sensitivity using a standard forced-choice procedure [25] . Sensitivity to bitter differed significantly between children and adults, with more supertasters among children and more non tasters among adults. No sex differences were observed for children, but among adults significantly less females were non tasters and more were supertasters The most common haplotypes in the Caucasian population, PAV and AVI recurred as common diplotypes PAV/PAV, PAV/AVI, AVI/AVI in 90.4 % of the subjects, the less common haplotype AAV was present in 8.7 % of the sample, the rare haplotypes (AVV, PVV, AAI) in 1% of the population, as expected.

Food Preferences, Age and PROP Status
Bitter and non-bitter tasting vegetables preferences (consumption and liking) were calculated for all age groups. As shown in Figure 2, both bitter and non-bitter tasting greens intake increased significantly with age (ANOVA; F=20.17, p<0.0001 and F=13,93 , p < 0,0001). As previously reported, no association was found between bitter and non-bitter tasting vegetables neither with PROP phenotypes nor haplotypes.

Children versus Mothers
In the mother-child dyads, the child was generally more sensitive to PROP than the mother. In fact, in 32% of the pairs the number of medium/ supertaster children exceeded that of mothers, while in only 15% of the pairs the mother was more sensitive than the child. In the remaining 53% of dyads, mothers and children shared phenotype irrespective of genotype. A comparison of the global preference score (consumption + fondness) of both bitter and non bitter tasting vegetables within the dyads revealed striking differences between the child and the mother (mean preference score= 9.7) for non-bitter tasting greens, and particularly for the bitter tasting greens (mean preferences score= 11.2). These differences decreased with the age of the child: adolescents showed a trend "more compatible" with that of their mothers (Pearson correlation r=-0.308 for non-bitter vegetables; r=-0.174 for bitter tasting ones; both P <0.05).
In addition, the TASR38 haplotypes affected the differences between children and mothers for non-bitter

Discussion
The main aim of this study was to evaluate in a Mediterranean population whether PROP genotypes/ phenotypes influence feeding behaviour in children and adults, and in mother-child dyads. Here we confirm our previous finding, obtained in a smaller sample [20], that PROP sensitivity differs between children and adults. In fact, the frequency of supertasters was higher in children, even in mother-child dyads sharing the same  Sensitivity to the bitter compound (tasting phenotype) was strongly related to the TAS2R38 haplotype, but the association between genotype and phenotype was more stringent for the PAV haplotype. TASR38 haplotypes affected the differences between children and mothers global preference score (consumption + fondness) for non-bitter and bitter tasting vegetables.  zinc ion concentration. However, Melis et al. [22] reported that, among PAV homo-and heterozygous, the supertaster status is not more frequently associated to the AA genotype. In line with the findings of Feeney et al. [28], and Tomassini Barbarossa et al., [29] we did not find a correlation between CAVI polymorphisms and the PROP phenotype.
In fact, the frequency of the gustin genotypes was not associated with the TAS2R38 genotypes or with the bitter sensitivity phenotypes. It is a taste-modifying gene, with no overall implication in producing the tasting phenotype.
PROP responsiveness is also linked to the secretion of salivary peptides from the basic proline-rich protein family. In fact, these peptides were found to be more abundant in supertasters than in non-tasters and their concentration in saliva was increased by PROP stimulation [30] ,which suggests that PROP sensitivity could be more complex than hitherto thought.  [49][50].
Regarding food preferences within the motherchild dyads, we show that the difference between mothers and children in terms of acceptance and consumption of greens decreased with the age of children, and it was significantly related to the children's TAS2R38 haplotype. Because only 68.9% of the mothers shared the same TAS2R38 haplotype with their children, these differences could explain a significant proportion of the discrepancy in food consumption between mothers and children. Consumption of vegetables depends largely on age and less on the specific sensing genotype. Within the mother-child dyads, the differences in food preference were related to age, and to the specific genotype of the child. Because mothers are not aware of the tasting genotype of their children, they may feel that their child's refusal to eat vegetables is a problem of relationship and behaviour and not of taste.
In fact, as shown in this and other studies [12,14,51,52], food preferences are also strongly